824 EEPOET— 1901. 



cliaracter confirms distiuctious ; as, for instauce, in separating tlie unitegminous 

 Betulese and (Joryleae from the bitegminous Querciue^. The explanation of all 

 these constructions may, I suggest, be sought for with better prospect of success 

 in the water-relationship and food-relationship of the integuments to the embryo 

 than in protective function and relations to pollination. It is, perhaps, not without 

 significance from this point of view that in, for instance, the Gamopetahc such 

 protective function as attaches to the tegumentary system in the seed is reduced 

 or extinguished through the development of indehiscent fruits, accompanied in many 

 Aggregatee and higher Ileteromerse by the sinking of the gynreceum in the torus, 

 and in many Bicarpelletse by its enclosure in a persistent accrescent calyx. 



All the information at our disposal seems to indicate that the tegumentary 

 system of the ovule is extremely adaptive, and that its characters are not of them- 

 selves of much phyletic import. An extended examination of its characters as an 

 organ of the nature I have depicted in relation to embryogeny is greatly needed. 

 It is made all the more interesting by the questions of development of endosperm 

 opened by the discovery of ' double fertilisation.' There is no more promising field 

 of investigation than this, for it must yield results infinitely more interesting 

 than the technicalities of formal morphology which have been for too long the 

 stimulus to ovular research. I am tempted to go further and to say that it might 

 supply an explanation of that most puzzling of subjects, the forms and curvature 

 of the ovule. The common assumption that these have relation to polUnation and 

 make the advent of the pollen-tube at the niicropyle easier is not altogether satis- 

 factory. For the curvature not infrequently seems to place the micropyle in a 

 position the opposite of favourable, and there is an absence of curvature in cases 

 where it would appear to be desirable. 



I will not dwell upon the subject of the seed itself as an advantage to the 

 Angiosperm. Its construction follows upon the successful water-relation pre- 

 viously secured. We all know how its manifold adaptations to dissemination 

 bring about its fortuitous deposition upon various soils, and the embryo is placed 

 well guarded within the seed-coat ready to take advantage of the moment when 

 moisture is sufficient for its germination. 



Whilst the seed-habit is the character which has primarily given to Angio- 

 sperms their advantage as a land-type,' their vegetative organs also show an 

 advance in their relationship to water upon those of the forms they have sup- 

 planted. I have already remarked that the growth-forms of the vegetation of the 

 present day are the same as those of old. That means that the early as well as 

 the later groups of vegetation have solved in much the same way, so far as general 

 form is concerned, the problem of the exposure in the atmosphere of a large 

 assimilating area with a sufficient mechanical support and adequate water-supply. 

 That wherever a water-carrying system is found in these growth-forms it dominates 

 the anatomy is witness to the importance of the water-relationships I wish to 

 emphasise. 



There are two features in the water-carrying system of Angiosperms in which 

 they are superior to the older types — namely, their general monostely and their 

 vasa. 



No one will contest that polystely is a less perfect mechanism for water-carriage 

 in a massive plant than is monostely. The limitation imposed by it to an incre- 

 ment in the area of carriage contrasts unfavourably with the openness in this 

 respect possessed by monostely. In the moister climatic conditions of the age of 

 domination of Pteridophytes polystely may have well sufficed for the water-needs 

 of the plants, especially of the dwarfer forms ; but even then, as we know, mono- 

 stely was the habit in many of the larger tree-foi-ms, and the development of a 



' Gymnosperms, sharing with Angiosperms the seed-habit, have in that had 

 advantage over Pteridophytes. Bat their flower-mechanism is much less perfect. 

 The reasons for their being bested as a class by Angiosperms must be complex. 

 Gymnosperms, as a whole as we know them, are less adaptive than Angiosperms. 

 The decadence of the cycadean line of descent may have been helped by their con- 

 servatism in the methods of water-carriage in the vegetative organs. The coniferous 

 type has held its own in the Northern Hemisphere. 



